Research map / Reinforcement learning / Policy gradient and actor-critic
Policy-Adaptive Estimator Selection for Off-Policy Evaluation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use adaptive importance fitting and subsampling instead of a non-adaptive heuristic | OPE estimator selection · Real-world e-commerce, evaluation policy j=1 | Relative Regret (e): 1.755 → 1.609 | Helped |
| Use adaptive importance fitting and subsampling instead of a non-adaptive heuristic | OPE estimator selection · Real-world e-commerce, evaluation policy j=2 | Relative Regret (e): 1.827 → 0.691 | Helped |
| Use adaptive importance fitting and subsampling instead of a non-adaptive heuristic | OPE estimator selection · Real-world e-commerce, evaluation policy j=3 | Rank Correlation (e): 0.015 → 0.124 | Helped |
| Use adaptive importance fitting and subsampling instead of a non-adaptive heuristic | Off-policy policy selection · Synthetic OPS, (β1, β2)=(-2.0, 2.0) | Relative Regret (p): 0.0341 → 0.0077 | Helped |
| Use adaptive importance fitting and subsampling instead of a non-adaptive heuristic | Off-policy policy selection · Synthetic OPS, (β1, β2)=(-2.0, 2.0) | Rank Correlation (p): 0.7452 → 0.9619 | Helped |
| Use adaptive importance fitting and subsampling instead of a non-adaptive heuristic | Off-policy policy selection · Synthetic OPS, (β1, β2)=(3.0, 7.0) | Relative Regret (p): 0.0325 → 0.0230 | Helped |
| Use adaptive importance fitting and subsampling instead of a non-adaptive heuristic | Off-policy policy selection · Synthetic OPS, (β1, β2)=(3.0, 7.0) | Rank Correlation (p): 0.778 → 0.972 | Helped |
Closest papers by meaning
- OPERA: Automatic Offline Policy Evaluation with Re-weighted Aggregates of Multiple Estimators (NeurIPS 2024 · similarity 0.91)
- Off-Policy Evaluation with Deficient Support Using Side Information (NeurIPS 2022 · similarity 0.90)
- Adaptive Estimator Selection for Off-Policy Evaluation (ICML 2020 · similarity 0.90)
- Benchmarks for Deep Off-Policy Evaluation (ICLR 2021 · similarity 0.89)
- Towards Assessing and Benchmarking Risk-Return Tradeoff of Off-Policy Evaluation (ICLR 2024 · similarity 0.89)
- Off-Policy Evaluation under Nonignorable Missing Data (ICML 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.